Objective: Depressive symptoms have been suggested to increase mortality risk but causality remains unproven. Depressive symptoms increase likelihood of smoking which is thus a potential factor modifying the effect of depressive symptoms on mortality. This study aims to assess if the association of depressive symptoms and all-cause mortality is affected by smoking.

Methods: A prospective cohort study in Finnish primary care setting was conducted among 2557 middle-aged cardiovascular disease (CVD) risk persons identified in a population survey. Baseline depressive symptoms were assessed by Beck's Depression Inventory (BDI) and current smoking by self-report. Data on mortality was obtained from the official statistics. Effect of depressive symptoms and smoking on all-cause mortality after 14-year follow-up was estimated.

Results: Compared to non-depressive non-smokers, the adjusted hazard ratio (HR) for all-cause mortality was 3.10 (95% CI 2.02 to 4.73) and 1.60 (95% CI 1.15 to 2.22) among smoking subjects with and without depressive symptoms, respectively. Compared to the general population, relative survival was higher among non-depressive non-smokers and lower among depressive smokers. Relative standardized mortality ratio (SMR) for all-cause mortality was 1.78 (95% CI 1.31 to 2.44) and 3.79 (95% CI 2.54 to 6.66) among non-depressive and depressive smokers, respectively, compared to non-depressive non-smokers. The HR for all-cause mortality and relative SMR of depressive non-smokers were not increased compared to non-depressive non-smokers.

Conclusion: Current smoking and increased depressive symptoms seem to additively contribute to excess mortality.